Swiss technology firm FinalSpark has introduced a program that offers scientists access to “biocomputers” composed of human brain cells. These biocomputers are made from clusters of lab-grown brain cells known as organoids. 

The concept of temporary living AI

FinalSpark says its biocomputers, or organoids, have a “lifespan” of up to 100 days. This relatively short operational window means that AI systems trained on these organoids may need to be periodically transferred to new clusters or face degradation.

Training these organoids involves twin setups: positive dopamine rewards and negative electrical stimulation. These methods help the organoids form new pathways more like a living human brain appears to learn. So far, these organoids are being watched around the clock to observe their behavior and find better ways to make them operate as they are told.

Universities jump on the opportunity to research new organoids

Roughly 34 universities have requested to use the new FinalSpark’s ‘biocomputers.’ However, the company has only agreed to the requests of 9 institutions, including the University of Michigan, the Free University of Berlin, and the Lancaster University in Germany.

The University of Michigan has said it will prioritize its research on the electrical and chemical prompts necessary to alter organoid activity and build on developing organoid-specific computer language. Meanwhile, scientists at Lancaster University Leipzig in Germany are trying to fit the organoids into different models of AI learning, enabling them to run on a larger scale and possibly gain ground over silicon-based AI systems.

FinalSpark’s biocomputers may already have some competition

The Swedish firm has ramped up its neuro platform facility, which is now capable of producing between 2000 and 3000 biocomputers. However, it may not be the only company producing ‘biocomputers’ for very long.

Ángel Goñi-Moreno, a researcher at Spain’s National Center for Biotechnology, said he is working on new cellular computers capable of responding to environmental stimuli. These computers could facilitate bioremediation and restoration of damaged ecosystems, a phenomenon he called “cellular supremacy.”

Moreover, Andrew Adamatzky of the University of the West of England focuses his work on fungi, which show similar electrical properties to those of neurons. He hopes to use them in a brain-like fungal computing system. Andrew said he has already successfully trained fungal networks to help computer systems perform certain mathematical functions. Andrew is hoping to scale up his research and make his fungi system able to learn and be used for reservoir computing and pattern.